Tube guitar amp: Hum and oscillation

[chr1s]

Hi there,

I spent the last days trying to build a guitar amp. At first the channels should have three and four tube stages, a cathode follower before the tone stacks and one stage for makeup gain for the power amplifier.

Starting with the high gain stage to test for stability, I get lots of hum and also oscillation, which can be controlled with the gain knob. The osscilation is independent from the master volume, it occures also if the tone stacks and later stages aren't connected.

The following images show the hum and the starting oscillation.
I rebuilt the stages, and testing with a sinus input (in this pictures the input is open with 33k to simulate a PU instead of being shortened to ground) it seems that the amp is not completely instabe (whith moderate gain) because the oscillation starts at certain points and goes off again.


I then rebuilt the amp to have three tube stages per channel. This way I got rid of the oscillation, but the hum remains.

I'm quite shure the problems are due to the chassis design and position of tubes/transformers...
The schamtic shouldn't be a problem, there are no fancy feedback loops, it's a straight on stage after another design. I don't have the schematic in electronic form atm, but if someone is interested I can make one.
It's very unlikely to get the amplifier running (I've quite given up with that), but I thought at least I can learn what I did wrong.


The chassis is a tub with the transformers to the left/right mounted in a wood case. The power amplifier is from a Dynacord Eminent II, the power transformer has 330V and 250V which result in ~400V for the preamp and screen grids and ~750V for the anodes. For the first two stages I used 4x75V zener diodes, providing ~300V.
The first two gain stages are at the middle tube, then left side of the right tube, both left tube stages, master volume, last stage and power amp.
It doesn't change much with the first stages at the left tube.

The rectifier is near the transformer, the voltage gets via (2) to the capacitors. (1) is the 750V, (3) and (4) the power/standby switches, (5) the input jack, (6) the point for star ground, (8) the chassis ground, (7) the zener diodes, (9) the 6,3V lines, (10) the power amp PCB and (11) the tone stacks.


In the following picture relais for channel switching and a supply therefore are included. They show the transformers (power left, output right)


Remarks:
- one thing causing some of the hum is mounting the in jack next to the standby switch. I changed that (switching wit gain pots).
- the star ground should be ok, although every tube stage only has one wire to ground.
- the cables are a mess (reason for oscillation?)

As I said, I will probably rebuilt the amp on another chassis, but I think it's a good example of how not to do it. So I hope at least I (and maybe others) can learn for the next time.
Also, as far as i remember on previous amps I didn't have oscillation that started at certain amplitude points und went of again (it rose from 'quiteness'). Is this a common thing and does this give any conclusion where the oscillation comes from?



I'm looking forward for any comments and suggestions about reasons for the hum and oscillation.


Chris

 

[W DeMarco]

From the looks of it you should try to clean up the wiring. All grounds should be as short as possible. I built an 18 watter that was a noisey little bitch. Once I really went through the wiring things got a whole lot better.

Once thats done go through all connections with her fired up and a chopstick. Tap on stuff... look for cold solders...(HIGH VOLTAGE you can die from a zap. BE CAREFUL. KEEP YOUR FREE HAND IN YOUR BACK POCKET) You might find some poor connections.

Make sure your heaters wiring is tightly coiled and try to route where there are few other interconnects.

Maybe get a proper chassis from an online amp builder site. Im sure you could find one that fits your needs.

Will

 

[W DeMarco]

Another warning is that your caps stay charged for quite a while. Don't start resoldering until you discharge them. I use some high voltage lights connected from B+ to ground then lower voltage lights to continue drain then finally I short direct to ground to be sure. No spark means you drained them good. Big spark = lifesaver.

 

[chr1s]

I forgot to mention, the hum is 50 Hz, the HT voltage (esspecially for the first stage) is very clean.
[quote author="W DeMarco"]
From the looks of it you should try to clean up the wiring. All grounds should be as short as possible. I built an 18 watter that was a noisey little bitch. Once I really went through the wiring things got a whole lot better.
[/quote]
You think with proper wiring the Amp could work fine?
Well it's at least worth a try.


thanks, Chris

 

[W DeMarco]

Its not that I think it will work fine. There could always be another source of probs but I know that Improper wiring in tube amps usually leads to noise.

 

[chr1s]

[quote author="W DeMarco"]Another warning is that your caps stay charged for quite a while. Don't start resoldering until you discharge them. I use some high voltage lights connected from B+ to ground then lower voltage lights to continue drain then finally I short direct to ground to be sure. No spark means you drained them good. Big spark = lifesaver.
[/quote]
The charge is no problem with this amp, I have a LED that shows when the 400V is below ~50V, and if I switch the amp off completly the voltage drops in a matter of seconds to ~5V. (grid bias is off->power tubes drain larger current)

Chris

 

[stanz]

Oscillations are almost always a lead dress issue.




http://ayanmusic.com/leaddress.htm



or go here, register, and look up the trainwreck builders guide in the files section.

http://ampgarage.com



Lead dress, where wires go and what other wires they are near or oriented toward, may actually be the most important part of building an amp. This would include grounding layouts and avoiding ground loops.

Kevin O'Conner has some excellent books. The Ultimate Tone #1 may be just what the doctor ordered.

www.LondonPower.com

 

[coatofarmseast]

As for Hum.... In fender amps and in general in guitar amps, the best place to connect grounds are not in a star organization.

Maybe try B+, screen grid, fillament, and bias ( if it is fixed bias) near the power transformer side, along with all other power amp section grounds. Then place all preamp grounds up on the input jack or close too it.

Is the input jack plastic cliff type or is it on the chassis?

Another trick I've seen is biasing the filaments above ground, say at the cathode if it is cathode biased.

Do you have balance on the filament voltage, if not try 2 100 ohm 2w R's to ground, you should do this anyway as protection. Maybe try DC for your high gain tubes.

So maybe a preamp on one side of the chasis, power amp on the other would be best, that is the way most fenders are.

On the note of oscillation, maybe a cap b/w the plates of the phase inverter or between the plates of your power tubes.

Hope this helps

Jesse

 

[babyhead]

Not my image, but I just have to say Hiwatts are the cleanest wired amps around.

Be careful.

 

[chr1s]

[quote author="stanz"]
Lead dress, where wires go and what other wires they are near or oriented toward, may actually be the most important part of building an amp. This would include grounding layouts and avoiding ground loops.
[/quote]
Never heard 'lead dress' before, so it should be a good idea to learn about it.
Thanks for the links.

@coatofarmseast:
The input jack is plastic. Some of the suggested things are already done. (Heater balance and above ground, but only 7 V)

On the note of oscillation, maybe a cap b/w the plates of the phase inverter or between the plates of your power tubes.

The oscillation occures also w/o signal connection to power amp or even tone stack. I tried a cap parallel to one anode resistance, but as experienced with other oscillations, highcuts usually don't solve such oscillations.

[quote author="babyhead"]
Not my image, but I just have to say Hiwatts are the cleanest wired amps around.
[/quote]
Well, there seems to be quite a difference... :green:


I think I will rewire the amp one more time properly and change to another chassis if there's still hum.

Chris